DE10339314B3 - Method for display control of different information in a vehicle and opto-acoustic information unit - Google Patents

Abstract

such Information units are used to display navigation and Media data used in modern vehicles and are out of practice especially with separate displays, but also with dual-view displays, on which the information is displayed alternatively in time, known. The display control is done manually there, which is negative on the comfort for the Driver and driving safety. In the method according to the invention therefore automatically becomes the current viewing direction of the driver (DRV) continuously detected and the driver relevant for the driver Information (NAV) only displayed almost instantaneously when the Driver (DRV) has eye contact with the dual-view display (DISP), whereas in the intervals without eye contact alternatively driving-unimportant Information (MEDIA) can be displayed, with the current line of sight through a video-based detection of head rotation and / or eye movement of the driver (DRV) based on a static or dynamic Detection method determined and in the evaluation only a binary decision (THR) is made, whether the driver (DRV) the dual-view display (DISP) looks current or not. It can be commercially available displays in conjunction with video cameras (CAM) and simplified software be used.

Description

The The invention relates to a method for display control of different information on a common dual-view display by the driver during driving in a vehicle, whereby also a possible passenger the dual-view display and an opto-acoustic information unit in a vehicle for displaying different information a dual-view display with a program unit and a display control unit, in particular for implementation of the procedure.

such Opto-acoustic information units are used, for example, for a representation of Fahrzeugführungs- and status information (navigation display for presentation of driving-relevant data, like road map, Traffic information, and route guidance information, Operating data about the Vehicle, such as speed and engine temperature) as well as for display of television pictures (media display for presentation not relevant to driving Data, for example video films, TV broadcasts, internet pages) used in modern cars and are different in practice embodiments known. There are also different methods of control the display on the display as an integral part of the optoacoustic Informa tion unit applied, mainly in their comfort for the driver as the main user. Although in the following often from the Application in the car or more generally in the traffic on the Country is assumed, the comments also refer to vehicles for the Traffic happening on and in the water and in the air. Everywhere over there, where with a vehicle found a way and information about the vehicle be collected and passengers must be transported Such information units to increase the comfort of the driver and its passenger can be used. However, the most common are currently still separate displays for used the navigation display and media playback. The media displays for the Passengers in the rear, for example, in the headrests of the front seats installed, folded out of the car roof or with the help of a flexible support arm designed as a satellite. For the passenger is the display either foldable in the area of the sun visor or retractable built into the instrument panel, with the risk that the driver's unobstructed field of view is restricted and the driver unintentionally is distracted by the media playback. The navigation display becomes common installed in the center console or below the rearview mirror placed on the instrument panel (this solution is often used in retrofittable systems ) Offered.

One completely other execution way is to have a common display for the temporally alternative (selective) navigation and media display (dual-view display, distinguished from the term "dual-view" for multimedia presentations in split-screen or picture-in-picture mode), which by additional optical and / or electrical components a selective reproduction the two sources of information in the direction of the driver or the Passenger allowed. This is the simplest form to a manual switch in an adapter cable, with the between the different information switched back and forth can. As a variant, information units are already known, Display the trip-relevant data only when the vehicle is at a standstill and while only let the sound continue if which accordingly uncomfortable for is the passenger.

From the DE 197 37 942 A1 is a permanently installed in the dashboard of a motor vehicle information unit is known in which the dual-view display has a liquid crystal element which is translucent in a first switching position in a large angular range and opaque in a second switching position outside a small angular range. The large angle range is focused on driver and front passenger, the small angle range, however, only on the front passenger. In the small angle range, the image shown is constantly visible, whereas the large angle range is dark in the off position. It is always only a picture content displayed, a switch between different image content must therefore be done manually. In order to obtain a particularly high difference between the light transmittances in the driver and passenger directions, the driving voltage of the liquid crystal element can be regulated as a function of the temperature. For this purpose, photosensors can be arranged inside and outside the small angular range, which detect the transmitted amount of light and optimize it via a change in the drive voltage. Due to the small angular range, the passenger, if he wants to watch a movie while driving, forced to fix his head strongly, which leads quickly to the fatigue, in particular by the occurring driving movements. About the method for display control, so to switch between the two areas of insight is not specified in the published patent application, but it is assumed that this is a manually operated electronic conversion that can be performed by the driver and the passenger , It must therefore in particular the driver to the possible consideration of the driving relevant data one perform active action, which requires appropriate movements and also time. This process is comparable with the manual switching on of a traffic transmitter on the radio. A distraction of the driver from the driving situation and a relatively large (possibly too large in a current decision situation) period from the desire or the requirement of data viewing up to their presentation thus occur. Furthermore, the switching of the various image contents must also take place constantly.

From the DE 199 20 789 A1 is a switching-free information unit known whose display works with a cylindrical lens grid, which allows the simultaneous viewing of two different image content from different directions. In principle, for this direction-selective playback similar methods are considered, as they are also applied for the selective reproduction of a left and a right stereo image in raster stripe shape in autostereoscopic 3D displays and are well known, so that at the same time dual viewing in the car to is a new application. The difference to the 3D rendering, however, is that the lateral distance between the two viewing zones is very large (about 70 cm, in the stereo playback, the lateral distance corresponds to the eye distance of about 65 mm), which makes appropriate demands on the screening , This embodiment has the general disadvantage that the local image resolution is reduced because only half of the pixels of the display are available for navigation display or for media playback. In the theoretically possible time-sequential direction-selective reproduction, the image repetition rate for both image contents would have to be halved. For this purpose, special displays with a very high refresh rate (more than 120 Hz) would be needed for the flicker-free reproduction of the two images (more than 60 Hz). With today's flat screens (LCD monitors, plasma screens, electroluminescent displays), this requirement can not be met.

In the DE 199 51 001 A1 a device is disclosed in which the direction of view of the driver is used to display information from different data sources in the field of view. The purpose of this device is to relieve the driver of the task of directing his gaze to a display device in order to read information. The information is displayed either in the driver's main field of vision (head-up display, in which case the windshield is used for reflection) or directly on the retina. By voice input or changes in the direction of view, the driver should also be able to activate the devices and functions displayed in this way in his field of vision. If the gaze is turned away from the road too long, a warning will be issued. The detected line of sight of the driver is not used to display information in the (main) field of view, but to automatically switch a display device that is in the secondary field of view (for example in the center console) when the viewing view is detected.

From the literature on computer user interfaces, the use of the viewing direction as a mouse replacement for the selection of actions or commands on a display is known (Eye Mouse, for example, is used as a solution for the physically disabled). In the DE 101 21 392 A1 This solution is applied in the vehicle to select a command displayed on a display with the gaze. The usual click button for activating the command on the steering wheel can be attached. There is thus a view-controlled selection or activation of elements on the display.

From the DE 197 46 764 A1 , from which the present invention proceeds as the closest prior art, a but also based on the cylinder lens grid principle, but switchable information unit is known in which a rasterized image content can be viewed from two different directions. This is done by the displacement of a pre-tensioned eccentric lever over which the cylinder lens grid plate for deflecting light in different directions is laterally displaceable. Although the resolution problem is reduced by the representation of only one image content in each case, the manual switchability that is associated with the comfort and time losses already mentioned above remains problematic. In addition, the displacement or pivoting of the cylinder grid plate requires only a change of the possible viewing angle, a switching of the image content must be made in addition. In summary, therefore, only methods and systems implementing them are known from the state of the art, in which the viewing angle and, independently of this, the displayed information can be set manually by the consciously implementable request of at least the driver. As a result, the attention of the driver is required to an increased extent and subtracted from the traffic and the comfort considerably reduced.

The object of the invention is therefore to be seen to provide a generic method for display control of different information on a common dual-view display, which offers the greatest possible comfort for the driver when viewing driving-related information. In this case, the driver should not be distracted by the unwanted consideration of the media playback of his task as a driver and thus a high driving safety can be guaranteed. Furthermore, however, even a possible passenger in his comfort when viewing journey-relevant data should not be burdened unnecessarily and comfort-reducing. The process should be fast and insensitive to interference in its course. Likewise, an opto-acoustic information unit for implementing in particular the method according to the invention should be simple and robust in its structure and in its handling. In addition, it should be, in particular the display used, inexpensive and accessible for current media presentations. The solutions for this task complex can be found in the two independent main claims. Advantageous Wei developments are shown in the respective subclaims and explained in more detail below in connection with the invention.

at the method according to the invention in dual-view mode, the current viewing direction automatically changes of the driver continuously detected. Is found that the Driver looks at the dual-view display, instantly become the driver driving-related information is displayed. This has the driver always the needed Information available, without performing actions got to. He is watched by the system and can make his decision to get driving-relevant information, intuitively through the Show choice of his line of sight. Investigations on driving safety have shown the driver a maximum of one second to read of the instruments available stands; if the attention is longer than one second, the accident risk increases significantly. The interruption of the media presentation for the Passenger will therefore in the application in tight time limits hold, so that the inventive method guaranteed a high viewing comfort for the passenger. Because every time the driver looks at the dual-view display, the driving-relevant data are displayed, there is also an educational process. The driver can provide non-driving data to a passenger in the car Phases of averted gaze are offered, do not look at. They disappear automatically when the driver sees the dual-view display look. The driver becomes visually through a media play not distracted and driving safety is guaranteed. Investigations with Navigation systems also have a systematic dependency the viewing frequency of the complexity of the current navigation demonstrated. Normally, the interruption of the media presentation during the View of the driver on the dual-view display (short-term switching in the navigation mode) no major impairment for the passenger; also usual film presentations on television are interrupted by much longer lasting advertising blocks. At a more frequent Interruption, the passenger will forego his media presentation anyway and watch the traffic.

• – Es können handelsübliche Flachbildschirme verwendet werden.
• – Das Dual-View-Display kann im üblichen Breitbildformat (4:3, 16:9,...) verwendet werden. Das ist besonders vorteilhaft, wenn für den Beifahrer TV-Programme oder Spielfilme von DVD dargestellt werden sollen (aufgrund der vertikalen Ausrichtung der Farbfilterstreifen ist für die bekannte Lösung mit einem Trennraster bei heutigen handelsüblichen Flachbildschirmen ein Betrieb im unvorteilhaften Hochformat notwendig).
• – Es ist keine besondere Bildsignalaufbereitung, wie beispielsweise spalten- oder zeilenweises Verkämmen der Bilder für Fahrer und Beifahrer, erforderlich.
• – Es gibt keinen Verlust an Bildauflösung und Helligkeit: Fahrer und Beifahrer sehen die volle Auflösung und Lichtstärke des Displays, beispielsweise 1024 × 786 Punkte.
• – Es gibt keine so gravierenden Realisierungsbarrieren wie in den stereobasierten Ansätzen aufgrund der dort zu realisierenden extremen Distanz zwischen den Betrachtungszonen.
• – Es wird durch den automatischen Verfahrensablauf ein optimaler Komfort für den Fahrer bei gleichzeitiger Gewährleistung der Fahrsicherheit geboten.

The almost delay-free presentation of driving-relevant data during eye contact of the driver with the dual-view display can be achieved by a corresponding detection technology. The driver is watched for him by a video camera without interference. The evaluation of the detected images is then binary as yes or no decision, since the system is only to decide whether the driver looks at the area of the display or not. As a result, data processing can be achieved in real time. A slight delay then results only by the control command for switching the image content. In the method according to the invention, therefore, both information sources are alternatively displayed using the full image resolution of the display, the switching between the two information sources takes place automatically as a function of the driver's current viewing direction. Due to the selective, non-simultaneous display of the different image content for driver and front passenger, the full local resolution of the display can be used in the usual orientation of the display (landscape format). The inventive method is thus technically easy to implement with commercially available components and shows the following advantages over the known approaches for a dual-view display summarize:

• - Commercially available flat screens can be used.
• - The dual-view display can be used in the usual widescreen format (4: 3, 16: 9, ...). This is particularly advantageous if the passenger TV programs or feature films to be displayed by DVD (due to the vertical orientation of the color filter strips is for the known solution with a separation grid in today's commercial flat screens operating in unfavorable portrait format necessary).
• - There is no special image signal processing, such as column or line by line combing the images for driver and front passenger required.
• - There is no loss of image resolution and brightness: driver and front passenger see the full resolution and light intensity of the display, for example 1024 × 786 points.
• - There are no such barriers to realization as in the stereobased approaches due to the extreme distance between the viewing zones to be realized there.
• - It is offered by the automatic procedure optimal comfort for the driver while ensuring driving safety.

In the realization of the method according to the invention it is provided that the optoacoustic Information unit in addition to the dual-view display, a program unit for the information to be displayed and a display control unit for carrying out the data processing requirements also has a video-based detection unit for continuous, automatic determination of the driver's current line of sight, at least one in the immediate vicinity of the display arranged video camera, which is aligned with the driver, and comprises an evaluation unit leading to the display control unit. Thus, it is a simple and inexpensive system that is also suitable for retrofitting. In this case, the dual-view display can preferably be integrated into the center console of a vehicle. The video camera can be arranged directly above the display in a screen. It is also possible to integrate directly into the display edge or - with transparent displays - to position it behind the screen.

By the automatic switching mode in the method according to the invention, the driver has no opportunity to look directly at distracting image content. However, there may be a risk of the driver following the media presentation in the peripheral field of vision and being distracted from observing the traffic. This can be prevented if the information is displayed in a direction-selective manner, with the driving-relevant information being displayed only to the driver and the driving-unimportant information only to the passenger. Such directional lighting for a dual-view display is for example from the DE 197 35 177 A1 known. For implementation, a backlighting system may be integrated with two alternative lighting directions, one of which is for the driver and the other for the passenger. To implement, for example, a commercially available transmission panel (LCD display) with modified backlighting can be used. The technical solution for direction-selective backlighting is preferably in two aligned to the driver or passenger white LED arrays. LEDs enable a nearly instantaneous activation / deactivation of the lighting, are also durable and have a high degree of efficiency. Upon activation of the driving-relevant data, the driver-oriented illumination device is activated at the same time. By contrast, in the phases of the media presentation, the lighting device aimed at the passenger is activated. In any case, this cleats the driver. However, the focused on this lighting device can also detect the passenger, so that he can also view the driving-related information.

through a provided manual operating device, for example a simple switch, to selectively enable, disable and Switching between the different information can be the Select co-driver, whether he is in the interruptions of his media presentation a dark screen or would like to be offered the navigation display. In addition, the display control also taken out of the automatic process and by the driver or passenger are configured manually in accordance with the situation. For example can the dual-view display permanently display the navigation information if no passenger Board is. For long highway ride with passenger, the navigation mode be temporarily switched off to prevent unwanted interruptions of film presentation to prevent. In the automatically controlled switching of the shown Picture content can about it Beyond that already be permanently included the driving-relevant Information will be displayed when the current line of sight the driver can not be determined. This can be, for example then be the case if the natural lighting the driver is not for an image capture is sufficient (twilight, Tunnel passage) or extremely unstable (drive through a tree alley with incident sunlight). The passenger must then a temporary one Interruption of his media presentation to accept. Special customer wishes regarding the switching behavior in extreme light conditions can considered in the software become. Even with unfavorable light and of course, of course however, during night driving, safe operation of the navigation display may occur guaranteed be illuminated when the driver actively, in particular with infrared light becomes. Therefore can be in the immediate vicinity the display of a driver-oriented lighting device, in particular with infrared light sources, be arranged. Especially can be provided that the at least one video camera and the driver-oriented lighting device behind one infrared light transmitting Disc, especially in the display area, are arranged. It can Thus, a complete system can be made commercially available, in which in the display edge the video camera and the infrared lighting already are integrated. This eliminates individual installations and disruptive cabling.

In addition to the integration of a direction-selective illumination system to increase traffic safety by reducing distraction of the driver can also be provided in the method according to the invention that the duration of the eye-view produced with the dual-view display detected and above a predetermined threshold, a warning signal from a acoustic and / or optical warning device is issued. From studies it is known that a motorist maximally for a period of one second to look at the navigation display or other instruments. However, if you look at the displays for a longer time, the risk of accidents increases significantly. By contrast, the abovementioned embodiment of the invention considerably reduces the risk of accidents in comparison to today's permanently switched-on navigation displays, because the driver is acoustically or optically (eg by a flashing display) asked to look again in the direction of travel after one second has elapsed. This measure can also be used in the context of a driver assistance system to prompt the driver by voice output to look forward, if he focuses too long on the navigation display.

For the software for the Method according to the invention with a video-based recognition gaze towards the dual-view display pixel-based methods (comparison of image patterns and edge patterns) and feature-based methods (finding and evaluating the position of the Eye angle, nostrils, Corner of the mouth etc. or the pupil and possibly additional light reflexes on the cornea) in question. The applied detection method is selected so that it as robust as possible is opposite changes the ambient lighting (driving in the sunshine or at night). To comply with the real-time condition, the algorithm used includes preferably few arithmetic operations, so one implementation on one Microprocessor platform possible is. This requirement also comes the binary decision scheme for the Umschaltautomatismus opposite. To be used also at night, is how already mentioned, Active lighting of the driver with infrared LEDs required. The evaluation is then necessarily based only on the use of intensity images without natural Color information.

to Clarification of the further software choice and the invention Based on figures in the specific description part of embodiments the operation of the method according to the invention in the automated Mode described. It shows the

1 an information unit installed in a car for carrying out the method according to the invention,

2 a block diagram of the basic principle of the method according to the invention,

3 a block diagram of the method according to the invention with additional camera,

4 Edge images as reference and current images,

5 Examples of motion vectors,

6 the measurement in the simplified corneal reflex method and

7 the calculation of the tolerance value for the simplified corneal reflex method

The 1 shows in a photomontage the arrangement of a gaze-controlled information unit VIU according to the invention with a dual-view display DISP in the center console MC of a car. Above the dual-view display DISP, infrared LEDs and a video camera are arranged behind an infrared-transmitting aperture IRB. Various manual mode switches MUD allow a choice of different operating modes, such as "Nav to permanent" NavOn, "automatic toggle" automatic and "Nav permanently off" NavOff) These switches can be used as simple mechanical buttons or as freely configurable touch screen elements in the display is integrated Other functional elements are arranged behind the dual-view display DISP and are not visible on the image.

in the Following are initialization and operation of gaze control as essential basic principles of the method according to the invention described. When natural Seeing are minor changes the viewing direction by a corresponding rotation of the eyeball in relation to the head.

For larger gaze movements or for a longer fixation period additionally takes place a rotation of the entire head in the direction of the fixed object. The known methods for gaze measurement have the goal of stably measuring the alignment of the visual axis, ie the current angle of rotation of the eyeball with respect to an external reference object, in the widest possible viewing angle range (corneal reflex method, cf. DE 199 53 835 C1 and the references cited therein) or minimal eye movements relative to the orientation of the head (limbus tracking method, cf. DE 199 53 835 C1 and the references cited therein). In the method according to the invention described here, in contrast to the known gaze measuring methods, it is a matter of making a yes / no decision with regard to the fixation of a selected object (the display) with the least possible outlay and a low susceptibility to interference. Therefore, the preferred embodiment of the implemented detection method may be based on the evaluation of a sum of externally visible and camera-to-be-detected image features indicative of gaze appeal. The basic principle for this is the reference image comparison (cf. 2 ). Before starting the journey, the driver switches the du al-view display with a pressure switch in the immediate vicinity of the display. At that moment, the camera mounted in the immediate vicinity of the monitor picks up an image of the driver, which is stored as a reference image for the visual attention to be detected. While driving, the reference image is compared with the current video images to reveal a gaze toward the dual-view display.

The image comparison determines a similarity value. This value can be calculated by comparing the intensity of the pixels in a predefined area (preferably a block of pixels comprising the area of the eyes). In a first step, the predefined area is searched by a block-matching method in the current camera image P (by comparison of the intensity values in image blocks of the reference image P _ref and the current camera image P). For the "best match", the accumulated absolute differences of the intensity values in the corresponding blocks of the current image P and of the reference image P _{ref are then combined} to form a similarity value be advantageous not to work with strongly illumination-dependent intensity values, but with image contours, using edge operators (eg the "Canny Edge" detector or the "Sobel" operator) to identify the significant contours in the reference image P _ref and in the current video image P Then, using the method described above, but based on the edge images, a similarity value is determined for the "best match." When the similarity value thus determined reaches a predefined similarity threshold (threshold THR), a signal is triggered which is the image source of the display DISP switches from the media source MEDIA to the navigation display NAV. At the same time, the rear lighting BL1 of the DISP display, which is aligned with the driver DRV, is switched on. As soon as the driver DRV turns his head / gaze away from the dual-view display DISP, the backlight BL1 is switched off and the media source MEDIA is switched on.

This extended to a directed backlighting basic principle of a gaze-controlled information unit VIU with a dual-view display DISP is In 2 shown as a block diagram. If the reference image P _ref and the current camera image P of the video camera CAM from the driver DRV sufficiently match (the difference is smaller than a threshold THR, yes / no decision), it is concluded on a Blickzuwendung. In this case, the navigation display NAV is turned on to the dual-view display DISP and the driver DRV-oriented backlight BL1 is turned on. The PASS backlight BL2 and Media Source MEDIA are temporarily switched off The procedures described below are designed to reduce the driver's attention to the dual-view display as reliably as possible from glancing at objects in the driving environment (when looking through the windscreen) or in the car (when looking at the rearview mirror and the remaining instruments) to distinguish. Particularly critical are those Blickzuwendungen that require a comparable lateral rotation of the head and eyes as the view of the dual-view display (same azimuthal orientation).

The first modified method involves a cross-check with an additional camera. An additional video camera CAM2 (compare 3 ) is preferably mounted below the inner rearview mirror in the same azimuthal position (relative to the driver's head) as the video camera CAM1 of the dual-view display DISP. During the initial calibration of the system, a reference _image P _ref1 when viewed in the direction of video camera CAM1 and a reference _image P _ref2 when viewed in the direction of video camera CAM2 are recorded. During operation, two similarity values SIM1 and SIM2 are determined by comparing the current camera images P1 / P2 with the associated reference _images P _ref1 / P _ref2 . The calculation is made for both signal sources according to one of the methods described above. A look at the dual-view display DISP is only concluded when the video camera CAM1 provides a better similarity value SIM1 than the video camera CAM2 (SIM1> SIM2) and this similarity value also reaches the predefined threshold for the similarity THR.

The 4 12 shows the edge images of the upper camera CAM2 arranged in the rear-view mirror region and the lower camera CAM1 arranged in the region of the dual-view display DISP (in each case the reference _image P _ref1 / P _ref2 on the _left and the current image P1 / P2 on the right, the driver DRV looks at the dual-view display DISP, that is, towards video camera CAM1. The drawn rectangles mark the block boundaries in the eye area used for the calculation of the similarity values SIM1 and SIM2. It can be seen that the upper blocks deviate significantly from each other and the lower blocks are very similar to each other. In this case, the method according to the invention signals a gaze to the dual-view display DISP and controls the corresponding switching.

Another method for improving the detection reliability is the evaluation of motion vectors ("visual flow") (cf. 5 ). The methods presented so far evaluate static image features for the detection of a possible viewing. The reliability of the method according to the invention, however, can be improved by taking into account the natural movement during the Blickzuwendung (dynamic Blickzuwendungsgeste by rotation of the head and eyes, movement of the eyelids). For this purpose, temporally successive images of the one or both video cameras are analyzed with methods of motion vector estimation (visual flow). Such methods are known, for example, in connection with digital image data compression (eg motion-compensated prediction in MPEG-2 data compression) and digital format conversion (motion-adaptive local / temporal image interpolation). There are already real-time hardware solutions for motion estimation - mostly based on a block matching technique. With known assuming geometry geometry (position of the video cameras, the dual-view display and the driver) and the assumption that the driver looks forward most of the time, can be closed only on the direction of the motion vectors on a possible Blickzuwendung. In conjunction with the evaluation of static image features, the result of the motion vector estimation can be used as part of a consistency check to improve the reliability of gaze application detection. The motion vectors tend to point downward (negative average of the vertical component of the vectors) as gaze is directed toward the center-console dual-view display.

Further detection methods are based on direct gaze measurement. With the methods described so far, eye movements can be difficult to detect without significant head rotation. Such Blickzuwendungen are especially to be expected if the dual-view display is not mounted in the center console, but higher (in or on the main instrument panel). In this case, a simplified form of the known corneal reflex method (cf. DE 199 53 835 C1 and cited therein). The simplified form provides that the light-emitting diodes required for generating the light reflection are arranged in the immediate vicinity of the camera axis. In this way, the pupil appears as a bright spot in the camera image (comparable to the red-eye effect when photographing with a flash unit); In addition, the light reflex when looking into the camera is in the center of the pupil image and can be localized there by an intensity threshold operation (cf. 6 ).

In the in 6 sketched embodiment, the dual-view display DISP is located directly above the gaze-measuring video camera CAM1. Infrared light emitting diodes IR-LED are arranged directly below the video camera CAM1. In the upper illustration in 6 the driver DRV looks directly into the camera or to the dual-view display DISP. The camera image P of the driver's eye EYE DRV shows a central reflex point on the cornea of the eye with iris and pupil. In the lower illustration in 6 The driver looks past the side of the dual-view DISP display. Clearly visible in the corresponding camera image P is the decentralized reflex point on the cornea.

To detect a Blickzuwendendung the dual-view display DISP can thus be given a largely Aufnahmgegeometrisch certain tolerance range in which the light reflection must be based on the center of the pupil when the driver looks at the dual-view display. In the 7 For example, the calculation of a tolerance value b 'for a given acquisition geometry is shown assuming the well-known Gullstrand's values for the average schematic eye. In contrast to the known corneal reflex method, therefore, no individual calibration of the measuring method is required for the application described here; instead, Gullstrand's average eye and easily measurable sizes of the technical design are used, which considerably simplifies the detection process. Furthermore, it is assumed for simplicity that the eye EYE turns around the center of curvature of the cornea (cornea). The sought tolerance value b 'is then calculated according to the set of rays b '= c * B / C * F / A.

Mean
A = distance of the measuring video camera (main plane) from the eye of the driver (eg 70 cm)
B = distance of the display edge from the optical axis of the video camera
C = distance of the display edge from the driver's eye (≅ A)
c = distance of the pupil plane from the center of curvature of the cornea (4.1 mm, according to Gullstrand)
F = distance of the (imaginary) image plane of the video camera from the lens center (≅ focal length, eg 4 mm) and
b '= distance of the light reflection from the center of the pupil in the camera image.

A

distance CAM / EYE

b '

tolerance

B

distance Edge DISP / CAM

BL1

backlight DRV

BL2

backlight PASSPORT

c

distance Pupil plane / center of curvature the cornea

C

distance Edge DISP / EYE

CAM

video camera

DISP

Dual view display

DRV

driver

EYE

eye the driver

F

distance the image level CAM / Lensenzentrum

IRB

infrared transparent panel

IR LED

Infrared light-emitting diode

MEDIA

media source

MC

center console

MUD

manual operating device

NAV

navigation display

P

Current camera image

P _ref

reference image

PASSPORT

co-driver

SIM

similarity value

THR

threshold for similarity value SIM

VIU

view controlled information unit

Claims (23)

A method for display control of different information on a common dual-view display by the driver while driving in a vehicle, wherein also a possible passenger can view the dual-view display, characterized in that automatically the current viewing direction of the driver (DRV ) is detected continuously and the driver-relevant information (NAV) is displayed almost instantaneously only when the driver (DRV) has eye contact with the dual-view display (DISP), whereas in the intervals without eye contact, information that is not relevant for driving ( MEDIA) can be displayed, wherein the current viewing direction determined by a video-based detection of the head rotation and / or eye movement of the driver (DRV) on the basis of a static or dynamic detection method and in the evaluation only a binary decision (THR) is made, if the Driver (DRV) currently displays the dual-view display (DISP) or not. Method according to claim 1, characterized in that that the information is displayed direction-selective, where the driving-related information (NAV) only to the driver (DRV) and the non-driving information (MEDIA) only for the passenger (PASS) are displayed. Method according to claim 1 or 2, characterized that permanently displays the driving-relevant information (NAV) if the current line of sight of the driver (DRV) is not can be determined. Method according to one of claims 1 to 3, characterized that the detection method used is pixel or feature-based is working. Method according to one of claims 1 to 4, characterized that as a static detection method the corneal reflex method for stable measurement of a possible huge Viewing angle range or limbus tracking method to capture a minimal relative eye movement is used. Method according to claim 5, characterized in that that is a simplified corneal reflex method, in which the pupil detected as a bright spot and a corresponding tolerance range for the Location of the light reflection defined on the cornea is used becomes. Method according to one of claims 1 to 6, characterized in that a difference-forming similarity comparison between a reference image (P _ref ), in which the driver (DRV) the dual-view display (DISP) looks, and a current image (P) performed and a similarity value (SIM) is determined, wherein the binary decision is performed as a function of a predetermined threshold value (THR) for the determined similarity value. Method according to claim 7, characterized in that that the similarity value (SIM) in a predefined area using a block matching method or with intensity values or edge operators. Method according to Claims 1 to 8, characterized that the dynamic detection method based on the evaluation of motion vectors, in particular using a block matching method. Method according to one of claims 1 to 9, characterized that the display control can also be done manually. Method according to one of claims 1 to 10, characterized that the duration of the dual view display (DISP) produced by the driver (DRV) Eye contact detected and above a predetermined threshold a warning signal, a warning message or a request to the driver is delivered. Method according to one of claims 1 to 11, characterized that the driver (DRV) is active, in particular with infrared light (IR-LED), is illuminated. Method according to claim 1 to 12, characterized in that the video-based detection is based on video recordings at two different recording locations, one recording location corresponding to the position of the dual-view display (DISP) and the other location azimuthally aligned above and from both recording locations from reference pictures (P _ref ). Opto-acoustic information unit in a vehicle to display different information on a dual-view display with a program unit and a display control unit, in particular to carry out of the method according to one of claims 1 to 9, characterized in that a video-based detection unit for continuous, automatic Determining the current direction of the driver (DRV) with at least one in the immediate vicinity of the dual-view display (DISP) arranged video camera (CAM) and an evaluation unit leading to the display control unit is provided are. Opto-acoustic information unit according to claim 14, characterized in that the dual-view display (DISP) in the center console (MC) of the vehicle is integrated. Opto-acoustic information unit according to claim 14 or 15, characterized in that the dual-view display (DISP) is designed as a flat display, in particular with landscape format. Optoacoustic information unit according to one of claims 14 to 16, characterized in that in the dual-view display (DISP) a switchable, the presentation of the different information serving backlight system (BL1, BL2) with two alternative directions of illumination, of which the one to the driver (DRV) and the other to the passenger (PASS) is integrated. Optoacoustic information unit according to one of claims 14 to 17, characterized in that a manual operating device (MUD) for selectively enabling, disabling and switching between the different information is provided. Optoacoustic information unit according to one of claims 14 to 18, characterized in that an acoustic and / or optical warning device is provided. Optoacoustic information unit according to one of claims 14 to 19, characterized in that in the immediate vicinity of the dual-view display (DISP) and / or the video camera (CAM) one on the driver (DRV) aligned illumination device (BL1), in particular with infrared light sources (IR LED) is arranged. Optoacoustic information unit according to one of claims 14 to 20, characterized in that the at least one video camera (CAM) and on the driver (DRV) aligned lighting device (BL1) behind an infrared transparent aperture (IRB), in particular in the display area, are arranged. Optoacoustic information unit according to one of claims 14 to 21, characterized in that azimuthally to the at least a video camera (CAM1) a second video camera (CAM2) is provided is. Optoacoustic information unit according to one of claims 14 to 22, characterized in that the second video camera (CAM2) in the area of the rearview mirror a car is intended as a vehicle.